Less Than Perfect
Designing real things inevitably creates flaws
I own an Oral-B toothbrush. I like it, especially its fat, curving handle, which fits my hand much better than the narrow straight handles of normal toothbrushes do. But there's a problem with the Oral-B: its distinctive handle is too big to fit into the holes of the toothbrush holder above my sink, so I end up balancing the toothbrush across the holder and regularly knock it into the sink or onto the floor.
In his latest book, the thoughtful and insightful Small Things Considered, Henry Petroski looks at the Oral-B toothbrush and other everyday objects to see what lessons they can teach us about the nature of design and the constraints that all engineers must face when bringing a new idea to fruition.
The designers of the Oral-B toothbrush were faced with a constraint that had influenced toothbrush design for decades: the placement in millions of homes of holders with one standard size and shape that manufacturers of bathroom fixtures had long since settled on. In the end, the Oral-B toothbrush people simply ignored the constraint, developing a toothbrush that, while perfect in one context (gripped in my hand when in use), is decidedly imperfect in another (the 99 percent of the time it sits in my bathroom).
Indeed, it is the external context in which a design must operate as much as its inherent merit that determines how perfect or imperfect it is. Petroski examines the paper cup, a staple of office water coolers and deli coffee stands. It became commercially successful only when the general public began to worry in the early 20th century about germs being transmitted by shared glasses and cups, which were far more durable and, at the time, easier to manufacture.
And without a doubt, it is context that drives the final nail into the coffin of perfect design. While the compromises demanded by the physical limits of materials, their cost, and how well the new design interacts with existing objects can all cause a gadget or process to fall short of perfection, the fact is that sooner or later, the evolving world in which an object exists will inevitably doom it to imperfection.
Petroski uses the humble telephone as an excellent example. When the rotary-dial telephone was introduced, it finally let callers connect to a phone number themselves instead of speaking to an operator. It reduced the costs and increased the reliability, privacy, and ease of use of the phone.
But--based in part on the utility of the rotary dial--as the telephone became more embedded into our daily lives the bloom went off the rose. Rotary dialing is slow, difficult for users with limited dexterity, and difficult to integrate with the advanced telephone services we take for granted today. The modern push-button touch-tone telephone allows us to store frequently used numbers, connect via digital networks, and do everything from checking our voice mail to booking airline tickets through automated systems.
It's not that the rotary-dial telephone had a bad design, but our expectations have changed, even of something as humdrum as a phone.
While some engineers may get depressed at Small Things Considered 's depiction of the impossibility of perfection, fearing that their profession is eternally doomed to failure, those who know what side their bread is buttered on will rejoice in the Sisyphean task. For one thing, it guarantees future employment for designers, and for another, it means that however great the achievements of the past, there will always be room for improvement--something to ponder while you're brushing your teeth.
Learning The Hard Way
When the report of the presidential commission charged with investigating the Challenger shuttle disaster was released in 1986, it became a classic study of engineering and managerial failure and was intensely examined in the aerospace industry and beyond. The 2003 report of the Columbia Accident Investigation Board should be scrutinized in similar fashion. Readers will be helped by the board's attention to creating a readable and well-presented report, and by its republication by Apogee Press. Apogee's version differs from the official government printing in being US $15 cheaper and published in a less bulky format.
Everyone in engineering and technical organizations should read this report, if only for the examination of how "engineering by viewgraphs" ultimately replaced nuanced technical discussion within NASA, contributing to the loss of Columbia and her crew. A CD-ROM that accompanies the report contains considerable supplemental material, including video of Columbia's ill-fated reentry and the smoking-gun tests that implicated a collision of foam with the shuttle's wing as the proximate cause of the disaster.